Report by BlackFog

47% of security leaders have experienced sensitive data exposure through AI tools

13 FINDINGSPublished Nov 4, 2026
View Original Report →

Key Findings

56% of organizations support their broader AI usage policies with comprehensive technical controls.

Technical ControlsAI Governance

70% of security decision makers in technology and telecoms say autonomous AI agents fall outside organizational controls.

TechnologyTelecommunicationsAutonomous Agents

39% of organizations report that AI adoption is moving faster than they can govern it.

AI GovernanceOrganizational RiskAI Adoption

Among organizations that say AI adoption is moving faster than governance, 27% have comprehensive AI controls in place.

AI GovernanceSecurity ControlsAI Adoption

32% of organizations can detect or monitor employees sharing proprietary design, creative, or marketing content but cannot consistently prevent it; 20% rely primarily on written policies or training to address that behavior.

Intellectual PropertyData Loss PreventionAI Tools

47% of security decision makers have experienced an incident in which sensitive data was exposed through an AI tool.

AI SecurityData ExfiltrationSensitive Data

Among organizations that experience sensitive data exposure through AI tools, 77% involve a company-approved AI tool.

AI SecurityData Exfiltration

35% of organizations have a written policy specifying what information employees must not enter into AI prompts but lack technical enforcement.

AI GovernanceData Loss Prevention

48% of organizations can provide records of the types of data that have left the organization through AI tools.

Data LossVisibilityAI Tools

56% of organizations say autonomous AI agents acting on behalf of employees fall outside existing controls.

Autonomous AgentsAI Governance

98% of organizations have an AI usage policy in place.

AI GovernanceAI Usage PolicyPolicy

62% of organizations can identify users or departments sending data to AI tools.

VisibilityData TrackingAI Tools

54% of organizations can block or restrict unsanctioned AI tools, and 32% can detect Shadow AI use but cannot consistently stop it.

Shadow AIAccess Control